1

Inappropriate drug prescribing

in older adults

Prof. dr. Mirko Petrovic

Department of Internal Medicine, Ghent University

Department of Geriatrics, Ghent University Hospital,

2

Structure

What does appropriate prescribing mean?

How should we identify subjects at risk of drug

related problems (DRPs) and adverse drug

reactions (ADRs)?

How should we review prescribing for an older

patient?

Which tools and strategies can help us to reduce

innapropriate presribing?

3

What is appropriate

prescribing?

A prescription that maximises efficacy and safety, minimises costs, and respects patient’s choices

Barber N. Pharm J 1996;257:289-91

« Pharmacological appropriateness » Only 1 dimension

Other dimensions What the patient wants

The « general good »

4

Prescribing more drugs than are

clinically indicated

Categories of inappropriate prescribing

MIS-

Inappropriate with regard to: Choice of drugs

Dosage

Duration

Modalities of administration

Drug interactions (/drug or /disease)

Cost

UNDER- Failure to prescribe drugs that are needed

OVER-

5

Is inappropriate presribing frequent?

Inappropriate prescribing

186 patients admitted to an acute geriatric unit

Almost 60% of prescriptions: 1 inappropriate rating

30% of patients were taking 1 drug-to-avoid

Under-prescribing in 50% of patients

Spinewine A et al. J Am Geriatr Soc 2007; 55:658–665

Condition

Osteoporosis

AF

IHD

HF

Myocardial

infarction

…

% patients undertreated

72%

40%: anticoag/aspirin

42%: aspirin

42%: ACEI

61%: b-blocker

…

Over- and mis-prescribing Under-prescribing

Consequences of inappropriate

prescribing

Drug related problems (DRPs) and Adverse drug

reactions (ADRs)

Drug interactions

Duplication of drug therapy

Decreased quality of life

Medication non-adherence

Unnecessary cost

Consequences of inappropriate

prescribing

Drug related problems (DRPs) and adverse drug reactions

(ADR) represent a major burden on health care

In Western countries ADRs cause 10-20% of all hospital

admissions, and are responsible for about 5-10% of in-

hospital costs

Lazarou J et al. JAMA. 1998;279:1200-1205.

Onder G et al. J Am Geriatr Soc. 2002;50:1962-1968.

Increased vulnerability to ADRs in older

people

Older people are 2-3 times more at risk for adverse drug reactions due to:

Pirmohamed M. BMJ 2004; 329:15-19

altered pharmacokinetics

altered pharmacodynamics

cumulative insults to the body (eg. co-morbidity, polyharmacy)

Barat I. Br J Clin Pharm 2001;51:615-622

non-compliance

lack of available data

medical errors

Tools and strategies to screen for and

prevent DRPs/ADRs

Medication review

Avoiding use op potentially inappropriate medications (PIM)

Computer-based prescribing systems

Comprehensive geriatric assessment (CGA)

Petrovic M et al. Drugs Aging 2012; 29: 453-462

Onder G et al. Age Ageing 2013; 42: 284-291

Screening- identification of

subjects at risk of ADR

Few data exist that allow stratification of patients

according to likelihood of an ADR

– An attempt to develop a risk stratification model: not enough

statistical power power to develop a risk score

Bates D et al. Arch Intern Med. 1999;159:2553-2560

– An attempt to identify specific patient’s characteristics

associated with an increased risk: restrospective study, relied

on voluntarily reported ADRs, under reporting

Johnston P et al. Am J Health Syst Pharm. 2006; 63:2218-2227

Development and Validation of a Score to Assess

Risk of Adverse Drug Reactions Among In-Hospital

Patients 65 Years or Older:

the GerontoNet ADR risk score

Onder G, Petrovic M, Tangiisuran B, Meinardi M, Markito-Notenboom W, Somers A, Rajkumar C, Bernabei R, van der Cammen T.

Arch Intern Med 2010, 170: 1142-1148

Odds Ratio 95% CI Points

4 co-morbid conditions 1.31 1.04 - 1.64 1

Heart failure 1.79 1.39 - 2.30 1

Liver disease* 1.36 1.06 - 1.74 1

No of drugs,

< 5

5-7

≥ 8

1

1.90

4.07

1.35 - 2.68

2.93 - 5.65

0

1

4

Previous ADR 2.41 1.79 - 3.23 2

Renal failure** 1.21 0.96 - 1.51 1

*transaminases > 2 x upper normal limit; ** GFR < 60 ml/min

Variables of the GerontoNet

ADR risk score

Screening- identification of subjects

at risk of ADR: evidence

The GerontoNET ADR risk score represents the only tool available so far to identify patients at risk of ADR, which may be target of interventions aimed at reducing their risk of ADR

However…

– it still should be validated in different settings and studies

– the need for identification of new risk factors to be added to the score

O’Connor M. et al. Age Ageing 2012;41:771-776.

Medication review

An individualized assessment provided by a clinical pharmacist: during which the medication list is analyzed in a structured manner, with full access to the medical file, in order to identify drug related problems.

First step: identification of all the medications that the patient is taking.

Second step: the medication list is screened for drug related problems i.e. any misuse, underuse or overuse of drugs.

Third step: possible solutions to the drug related problems (DRPs) are then discussed with the treating physician and, if possible, with the patient.

Medication review

Level 0

AD-HOC

Unstructured,

opportunistic

Level 1

PRESCRIPTION

REVIEW

Level 2

TREATMENT

REVIEW

Level 3

CLINICAL MEDICATION

REVIEW

Technical review of

list of patient’s

medicines

Review of

medicines with full

patient’s notes

Face-to-face review

of medicines and

condition

Reporting adverse drug reactions on a geriatric

ward: spontaneous reporting vs. patient interview (by

pharmacists)

Patients Patients with ADRs Number of

ADRs

spontaneous reporting 168 12 12

patient interview 56 23 32

Interviewed patients (n = 56)

Mean Median Range

Age (years) 80.1 80.0 62 – 94

Length of stay (days) 18.6 14.0 4 – 61

Number of drugs

patients with ADR (n=23) 9.3 8.0 6 – 16

patients without ADR (n=33) 8.3 9.0 3 – 14

Somers A et al. Eur Clin Pharmacol 2003;58:707-14

Reporting adverse drug reactions on a geriatric

ward: spontaneous reporting vs. patient interview (by

pharmacists)

Gender Male 10 (43%)

Female 13 (57%)

Causality Probable 23 (72%)

Possible 9 (28%)

Level 1 = no change 13 (41%)

2 = stopped / dose changed 12 (37%)

3 = stopped + additional therapy 7 (22%)

Severity Serious ADR 12 (38%)

Non-serious ADR 20 (62%)

Type Type A 32 (100%)

Type B 0 (0%)

Results of the patient interview

Somers A et al. Eur Clin Pharmacol 2003;58:707-14

19

Impact on appropriateness of prescribing

Spinewine et al. Lancet 2007;370:173-84.

RCT, 203 patients, one acute geriatric unit, Belgium

Pharmaceutical care from admission to discharge

- ↑ appropriateness of prescribing (MAI, ACOVE)- 90% acceptance rate

- Trend toward mortality and ED visits

RCT, 400 patients ≥80y, 2 internal medicine wards, Sweden

Pharmaceutical care from admission to discharge(+ after)

- 16% hospital visits- 46% ED visits- 80% drug-related readmissions

Spinewine A et al. J Amer

Geriatr Soc 2007; 55:658-65Gillespie U et al. Arch Intern Med

2009;169:894-900

Impact on appropriateness of prescribing

Medication review: evidence

Good evidence that collaboration with pharmacists

can decrease the risk of drug-related problems

Mixed / lacking evidence for effect on:

– Health outcomes

– HRQoL

– Cost-effectiveness of care

Chisholm-Burns Med Care. 2010;48:923-933

Spinewine et al. Drugs Aging. 2012;29:495-510.

Avoiding use of potentially

inappropriate medications (PIM)

Medication Assessment Tools

1) Explicit (criteria based): drugs to avoid– Beers (1991, updates 1997, 2003, 2012, 2015)

– McLeod (1997)

– ACOVE: Assessing Care of Vulnerable Elders (2001)

– IPET: Improved Prescribing in the Elderly Tool (2002)

– STOPP: Screening Tool of Older Person’s Prescriptions/ START: Screening Tool to Alert doctors to Right Treatment) (2008, 2015)

2) Implicit (judgement based):

– MAI: Medication Appropriateness Index (1992)

– GMA: Geriatric Medication Algorithm (1994)

– Lipton’s criteria (1993)

24

Avoiding use of potentially inappropriate

medications (PIM)

Explicit

– criterion-based

– reviews, consensus,

experts

– focus on

drugs/diseases

Process prescription accords

with accepted standards

should have causal links to important outcomes

Implicit judgement-based

focus on the patient

Outcome indicators of

adverse outcomes

25

Explicit

Process

Implicit

Outcome

- LA-BZD

- LA-BZD in

patients with fall

Patient with LA-BZD for

insomnia for 5 years,

other risk factors for fall,

patient open to attempt

progressive

discontinuation

Admission to hospital for

fall and patient taking a

LA-BZD

Avoiding use of potentially

inappropriate medications (PIM)

26

Beers criteria

Developed by expert group in US in 1991, updated in 1997, 2003, 2012, 2015

Drugs with risk > benefit in older patients

2 parts:

– non-recommended drugs in older people

– conditionally non-recommended drugs

– O/M

Beers et al., Arch Int Med 1991;151:1825-32; 1997;157:1531-36; 2003;163:2716-24; JAGS 2012;60:616-31

Examples of non-recommended drugs in older

people

Generic name

Propoxyphene

Indometacine

Phenylbutazone

Pentazocine

Oxybutynin

Flurazepam

Amitriptyline

Perphenazine-amitriptyline

Doxepine

Meprobamate

Lorazepam > 3 mg daily

Oxazepam > 60 mg daily

Alprazolam > 2 mg daily

Temazepam > 15 mg daily

Zolpidem > 5 mg daily

Generic name

Triazolam > 0,25 mg daily

Diazepam

Digoxine > 0,125 mg daily

Dipyridamole

Methyldopa

Chlorpropamide

Belladonna alkaloids

Chlorpheniramine

Diphenhydramine

Hydroxyzine

Cyproheptadine

Promethazine

Dexchlorpheniramine

Ergot mesyloids

Iron suppelements > 325 mg daily

Ticlopidine

Conditionally non-recommended drugs in older

people

Pathology treatment

Heart failure sodium containing drugs (sodium

carbonate: effervescent tablets)

Gastric ulcer NSAIDs, aspirin

Arrhythmia Tricyclic antidepressants

Depression methyldopa

Obesity Olanzapine

29

Beers criteria

Somme drugs controversial

Many drugs not available in Europe

Better situation with the 2012 version

Only 2 aspects of inappropriate prescribing

Beers et al., Arch Int Med 1991;151:1825-32; Arch Int Med 1997;157:1531-36 and 2003;163:2716-24

☹

☺ Easy and rapid to use

30

Beers criteria

Dalleur O et al. JAGS 2012;60:2188-2189

31

ACOVE criteria

Assessing Care Of the Vulnerable Elders

Literature study + expert opinion

22 pathologies, syndromes, clinical situations

236 indicators (prevention, diagnosis, therapy, monitoring)

68 medication-related indicators

If… then… (unless…)

– O/U/M

Wenger N et al. Ann Intern Med 2001;135:642-6

32

ACOVE criteria

• Continuity of care

• Dementia

• Depression

• Diabetes mellitus

• End-of-life care

• Falls and mobility disorders

• Hearing impairment

• Heart failure

• Hospital care

• Hypertension

• Ischaemic heart disease

• Malnutrition

• Medication management

• Osteoarthritis

• Osteoporosis

• Pain management

• Pneumonia and influenza

• Pressure ulcers

• Screening and prevention

• Stroke and atrial fibrillation

• Urinary incontinence

• Vision impairment

Domains of care taken into consideration

Wenger N et al. Ann Intern Med 2001;135:642-6

Indication Therapy

Diabetes low dose aspirine

Diabetes with proteinuria ACE inhibitor

Diabetes, TC > 240 g/dl lipid lowering agent

Heart failure with LVEF < 40% ACE inhibitor

Therapy Monitoring

Digoxine, with symptoms of intoxication determine plasma level within 24 hours

Anticoagulans determine INR 3 days after initiation

Anticoagulans (chronic use) determine INR monthly

Therapy 1 Therapy 2

Corticoids during > 1 month Start Calcium + Vitamine D

NSAID and hystory of peptic ulcer PPI or misoprostol

Criteria concerning therapy

Criteria concerning monitoring

Criteria concerning

prevention

34

ACOVE criteria

Operationalisability

No recent update

Wenger N et al. Ann Intern Med 2001;135:642-6

☹

☺ Geriatric conditions included

Encompass Tx, prevention, monitoring, education

and documentation

Applicable to patients with dementia and poor

prognosis

35

STOPP / START criteria

Screening tool of older persons’ potentially

inappropriate prescriptions (STOPP)

80 criteria, O/M

Screening tool to alert doctors to the right treatment

(START)

34 criteria, U

O’Mahony D et al. Age Ageing 2015;44:213-8.

STOPP/START criteria

STOPP

Aspirin > 150mg/d

SSRI in case of clinically significant hyponatremia

PPI for an ulcer in full therapeutic dose > 8 w

START

- Antidepressant in case of major depressive symptoms

during at least 3 months

Avoiding use of PIM: evidence

Use of STOPP/START criteria leads to significant and

sustained improvements in the appropriateness of

prescribing at discharge and for up to 6 months after

dischargeGallagher P et al. J Clin Pharm Ther. 2007; 32:113-21.

STOPP criteria seem significantly associated with

avoidable adverse drug events that cause or contribute

to urgent hospitalizationGallagher P et al. Clin Pharmacol Ther 2011;89:845-854.

Beers vs STOPP START

Similarities

Criteria: BZD & falls; CCB & constipation; long-acting

sulfonylurea;…

Differences

75% of Beers criteria do NOT overlap with STOPP

55% of the STOPP criteria are not part of Beers

Beers: more focus on anticholinergics; delirium; dementia

STOPP: more focus on anticoagulants; opiates; PPIs

38Dalleur O et al. JAGS 2012;60:2188-2189

Beers vs STOPP START

39Dalleur O et al. JAGS 2012;60:2188-2189

40

Explicit instruments

Pros of using explicit criteria in your daily

practice

Relatively easy to remember and to detect

Provides support to identify inappropriate prescribing

in the elderly

HOWEVER…

41

Explicit instruments

Cons of using explicit criteria in daily

practice

This is just one part of the story…

The patient’s perspective is often not taken into

consideration

We should not limit our evaluation to the application

of such criteria

42

Medication Appropriateness

Index (MAI)• 10 questions per drug

Hanlon et al. Am J Med 1996;100:428-437

1. Valid indication?

2. Appropriate choice?

3. Correct dose?

4. Modalities of treatment correct?

5. Modalities of treatment practical?

6. Clin. significant drug-drug interactions?

7. Clin. significant drug-disease interactions?

8. Duplication?

9. Appropriate duration?

10.Cost?

44

MAI

Time consuming

Knowledge-dependent

Hanlon et al. Am J Med 1996;100:428-437

☹

☺ Comprehensive and systematic

Includes operational definitions, explicit

instructions, and examples

Excellent as an educational « tool » for students!

Computer-based prescribing systems

Clinical Decisions Support Systems (CDSS) and Computerized Prescription Support System (CPSS) are interactive softwares, designed As potentially powerful tools to prevent ADRs

To support at the time of prescribing

All categories of inappropriate prescribing can be addressed, if prescription data are linked to clinical data

Computerized Provider Order Entry Systems (CPOE), which are based on these softwares, enable providers to enter medical orders into a computer system that is located within an inpatient or ambulatory setting.

Schiff G et al. JAMA 1998; 279: 1024-9.

Translating Quality Measures into

Clinical Decision Support

Co

mp

lex

ity

Validity

Drug

Data

Drugs & Dx’s

Drugs, Dx’s

& Labs

Drugs, Dx’s, Labs

& Clinical Info

Computer-based prescribing

systems

Disadvantages

Very few studies demonstrated an improvement in patient

outcomes

Challenging to implement

Existing systems are not geriatric specific

High volume of alerts: risk of unimportant warnings

Some prescribers are reluctant to use

Gurwitz J et al. J Am Geriatr Soc 2008; 56: 2225-2233.

Wolfstadt J et al. J Gen Inten Med 2008;23:451-458.

Strom B et al; Arch Intern Med. 2010;170:1578-1583.

Comprehensive geriatric assessment

(CGA)

Medical complexity plays an important role in the onset of ADR and should always be considered before prescribing a pharmacological treatment in older people.

Drugs that have proven in clinical trials clear beneficial effects to treat a chronic conditions and which use is indicated in clinical guidelines should be used carefully in complex older adults – since they may interact with co-existing diseases or

geriatric syndromes, may not be assumed correctly because of presence of cognitive deficits, disability or social problems or may be useless because the health expectancy of the patient is too short to determine a beneficial effect of the drug.

Tinetti M et al. N Engl J Med 2004; 351: 2870–74.

Comprehensive geriatric

assessment (CGA): evidence

CGA in association with a multidisciplinary team

(assessing and managing the health care problems

identified by the CGA, and developing individualized

care plans) results in more detailed evaluation,

improved care planning, and overall better quality of

care. Ellis G et al. BMJ. 2011;343:d6553.

Limitation: heterogeneity in terms of structural components and

care processes.

Comprehensive geriatric

assessment (CGA): evidence

CGA allows a complete and global assessment and management of the health care problems, including evaluation of drugs with the goal of recognizing and preventing potential drug-related problems and improve quality of prescribing.

Onder G et al. Curr Drug Metab 2011;12:647-651.

• CGA associated with a multidisciplinary team approach, as compared with usual care in frail older adults shows a 35% reduction in the risk of a serious ADRs and a substantial reduction in unnecessary and inappropriate drug use.

Schmader K et al. Am J Med. 2004;116:394-401.

Garfinkel D et al. Arch Intern Med. 2010;170: 1648-1654.

Systematic approach for drug cessation

in complex older adults

52

COLLABORATIVE CARE

Multidisciplinary teams

Geriatric medicine

services/CGA

Collaboration with

Clinical pharmacists

Nurses

Collaboration with the patient

Computerized support

Educational approaches

THM: Conclusions

None of the existing approaches shows a clear beneficial effect on patients’ health outcomes: available evidence on the impact of medication review, avoidance of PIM, computer-based prescribing systems and CGA is mixed and controversial.

A main limitation of all the described approaches is the lack of standardization.

Large differences are described in the delivery of the pharmacist-led medication review.

Criteria to assess quality of prescribing vary across countries and no widely accepted gold standard exists, yet.

Computer-based prescribing systems are often home-grown and they implement different types of information, tools and algorithms.

Geriatric assessment and management programs are heterogeneous in terms of structural components and care processes.

THM: Conclusions (cont.)

Most of the available research is focused on a singleintervention targeting either clinical or pharmacological factors causing ADR.

When these approaches were combined- as for studies assessing the efficacy of an intervention based on experienced pharmacists performing medication review in the context of a multidisciplinary team- positive effects on patients’ health outcomes were shown.

Safe drug use goes along with global assessment of patients clinical and functional parameters and that integration of skills from different health care professionals is needed to address medical complexity of older adults.

The challenge for future research is to integrate valuable information obtained by existing instruments and methodologies in a complete and global approach targeting all potential factors involved in the onset of ADR.

SENATOR is a Collaborative Project funded by the European Commission under the 7th Framework Programme

http://www.senator-project.eu

Development and clinical trials of a new Software ENgine for

the Assessment & Optimization of drug and non-drug Therapy

in Older peRsons

IGRIMUP

Invitation to join

IGRIMUP (International Group for Reducing Inappropriate Medication Use & Polypharmacy), founded during IAGG 2013 in Seoul, Korea is an open group focused on preventing negative effects of drugs by avoiding polypharmacy and Inappropriate Medication Use (IMU).

Researchers capable of promoting this goal in their country and/or internationally are welcome. Please feel free to forward us names of other leading figures, regardless of country or profession, that are interested in and willing to join the group.

We hope that you will share your thoughts about the issues prevalent in your country’s health care system on reducing inappropriate medication use, and ideas for change.

Doron Garfinkel, MD – dgarfink@netvision.net.il

Graziano Onder, MD – graziano.onder@rm.unicatt.it

Mirko Petrovic, MD – mirko.petrovic@ugent.be